EP2293039A1 - Method for quality control of dynamic behaviour of a railway vehicle - Google Patents

Abstract

The method involves extracting remarkable values of characteristic representing imperfection of track e.g. raising of track, from a database, and matching the remarkable values of measured variable with characteristic cells of the track. A graph, showing the remarkable values of the measured variable representing behavior e.g. acceleration, of rail transit car, is established based on considered imperfection characteristic, for analysis section belonging to same statistic class. Values obtained in the graph are compared with acceptable limited value.

Description

The field of the present invention is that of the railways and more specifically that of the qualification tests of railway vehicles.

The various rolling stock intended for use on railway tracks are subject to evaluations to obtain a certification allowing them to circulate on the network of one or more of the countries of the European Union. The aim of this certification is to guarantee the safety of passengers and railway agents with regard to the risks inherent in this mode of transport and in particular the risk of derailment.

To ensure the safety of the rolling stock tests are first made in static to measure various parameters such as the stiffness of the damping systems and more generally the transfer function of the vibration transmission. Dynamic tests, in real conditions of traffic on rails, must also be carried out, during which various parameters such as the level of the vibrations, the value of the acceleration experienced by the car body or the bogies, the force of contact between the wheel and the rail, etc. These tests are conducted under average conditions representative of normal use and under extreme conditions to judge the quality of the rolling stock, and then compared to the values set in the corresponding certification standard.

The results of these tests, however, are extremely dependent on the quality of the path on which they are made. Optimally they should be carried out on a test track whose characteristics are well known. However, for reasons of cost and lack of availability of such channels, tests are carried out on existing channels, which are used on a regular basis for commercial traffic and which are made available for testing only occasionally and affected. this task only in relatively short deadlines. These paths are obviously imperfect and more or less adapted to the conduct of qualification tests. Their precise characteristics are also known insufficiently precise to perform qualification tests in the best conditions. Patents or patent applications, such as documents US 5986547 , FR 2741026 , US 5775230 , WO2005 / 036199 or US 5433111 , describe in this regard, track quality detection devices, associated with odometry resetting devices, primarily to ensure the capability of a track to circulate trains. They do not test a vehicle to ensure its ability to flow.

It is not possible, for the teams in charge of the tests, to carry out, beforehand, a precise cartography of the track on which the test will be carried out. This operation would in any case be long and extremely expensive and is therefore not practically feasible.

It is therefore necessary to find a simplest possible method of analyzing the quality of the rolling stock, which makes it possible to take into account the imperfections of the track on which the test takes place and to discriminate the phenomena which come from the track of those coming from the vehicle under test. It must make it possible to measure, in real time during the test, the corrections to be made to the reference system of the characteristics of the track.

One of the essential characteristics that must be corrected lies in the odometry of the track, that is, in the knowledge of the distances of the various remarkable points on the track, with respect to a point chosen as the origin. This corrected odometry must in particular accurately give the kilometer distances of entry and exit curves of the track.

The databases that characterize a route divide it into a series of sections, which are either straight sections, referred to as alignments, or circular curve sections, or finally so-called connecting sections, which provide the necessary transition between the sections. sections in alignment and the sections in curve, at the radius of curvature and the slope of the track, the cant being the difference in height of one rail relative to the other. All these rectilinear and curved sections are listed and, for each of them, is indicated, among others, the odometrical distance of entry in the section, the odometrical distance of exit, the radius of curvature on this section and the slope applied to the way. These databases include a number of errors, due to, among other things, poor positioning of geographical boundaries (or kilometer points Pk) which serve as reference points for drivers and which are installed on the side of the track, or drifts, in time, in the physical constitution of the way.

Methods have been developed to carry out a test by overcoming, on the one hand, the lack of knowledge of the actual odometry of the track on which test will be carried out and, on the other hand, the possible difference between the reference systems due to the different odometry systems used to measure distances. A manual method is thus known which consists in classifying the various sections into statistical geographical categories and then comparing the measurement results made during a test with the ground data corresponding to the statistical category of the section. We understand that this method is tedious to implement. There is also known a method of placing along the path of the synchronization terminals so as to mark the actual inputs and outputs of the curves of the channel. This method is, as already indicated, long and expensive and poorly suited to conducting tests on a lane in service assigned late to the test to be conducted.

The object of the present invention is to remedy these drawbacks by proposing an extremely simple method, even on an imperfectly known route, of evaluating the quality of a railway vehicle, which does not require expensive means, which can be implemented operates automatically and is not sensitive to inaccuracies on the characteristics provided by the reference database of the railway on which the evaluation test or qualification of said vehicle takes place. It further comprises a method for resetting the odometry of the track.

• a. découpage de ladite voie en tronçons d'analyse,
• b. classement desdits tronçons d'analyse en classes statistiques regroupant les tronçons présentant des caractéristiques géométriques analogues,
• c. recalage, par une méthode connue, des distances odométriques des divers tronçons d'analyse avec les distances odométriques des tronçons figurant dans la base de données,
• d. mesure, lors de la circulation du véhicule sur la voie, d'au moins une grandeur (γ) représentative du comportement dynamique du véhicule, en réponse aux imperfections de la voie
• e. dépouillement de ladite grandeur, dans chacun des tronçons d'analyse parcourus par le véhicule, et extraction des valeurs remarquables de cette grandeur sur chaque tronçon,
• f. extraction, à partir de la base de données, des valeurs remarquables d'une caractéristique représentative des imperfections de la voie, dans chacun des tronçons d'analyse parcourus par le véhicule,
• g. appartement, sur chaque tronçon, des valeurs remarquables de la grandeur mesurée avec celles de la caractéristique de la voie,
• h. établissement d'un diagramme donnant les valeurs remarquables de ladite grandeur en fonction des valeurs remarquables de la caractéristique d'imperfection considérée, pour tous les tronçons d'analyse appartenant à une même classe statistique, et
• i. comparaison des valeurs obtenues dans ce diagramme avec une valeur limite admissible.

To this end, the object of the invention is a method of evaluating the conformity of the dynamic behavior of a railway vehicle with respect to the stability rules of rail vehicles, by means of a test comprising a circulation of said vehicle. on a railway track, said track being described in a reference database by indications on its geometry identified with reference to an odometric distance from an origin point, said database furthermore comprising information on the imperfections of said track located also by their odometric distance from the origin point. This process is characterized by the fact that it comprises the following steps

• at. dividing said channel into analysis sections,
• b. classifying said analysis sections into statistical classes grouping the sections having similar geometrical characteristics,
• vs. resetting, by a known method, the odometric distances of the various analysis sections with the odometric distances of the sections contained in the database,
• d. measuring, when the vehicle is traveling on the track, at least one magnitude (γ) representative of the dynamic behavior of the vehicle, in response to the imperfections of the track
• e. detachment of said quantity, in each of the analysis sections traversed by the vehicle, and extraction of the remarkable values of this quantity on each section,
• f. extracting, from the database, the remarkable values of a representative characteristic of the imperfections of the track, in each of the sections of analysis traversed by the vehicle,
• boy Wut. flat, on each section, remarkable values of the quantity measured with those of the characteristic of the track,
• h. drawing a diagram giving the remarkable values of said quantity as a function of the remarkable values of the imperfection characteristic considered, for all the analysis sections belonging to the same statistical class, and
• i. comparison of the values obtained in this diagram with an admissible limit value.

In a particular embodiment, the representative characteristic of the imperfections of the track is the training of said track.

In another embodiment, the representative characteristic of the imperfections of the track is the spacing of the rails of said track.

In yet another embodiment, the representative characteristic of the imperfections of the track is the leveling of said track.

Advantageously, the method further comprises a step of repeating steps f to i for each characteristic representative of an imperfection of the channel contained in the reference database.

Preferably, the method also comprises carrying out steps a to j for several speeds of passage of the vehicle on the track and a step of determining the maximum authorized speed in commercial service on a track as a function of the speed for which the value maximum imperfections presented by said path corresponds to the achievement by a representative magnitude of the vehicle behavior of its limit value.

In a particular embodiment, the division of the channel into analysis sections is performed as a function of the radius of curvature of said sections.

Preferably, the statistical classes are defined according to intervals of the value of the radius of curvature of said analysis sections.

In another embodiment, the division of the channel into analysis sections is performed as a function of the value of the cant of the track on said sections.

Preferably, the statistical classes are defined according to intervals of the value of the cant on said analysis sections.

The invention will be better understood, and other objects, details, features and advantages thereof will appear more clearly in the following detailed explanatory description of an embodiment of the invention given as a purely illustrative and non-limiting example, with reference to the accompanying schematic drawings.

• la figure 1 est une représentation schématique d'un véhicule en essai sur une voie ferrée,
• la figure 2 est une représentation schématique, en vue de dessus, d'une voie ferroviaire sur laquelle se déroule un essai de qualification.
• la figure 3 est une représentation schématique, en fonction de la distance odométrique, d'une information (ici, le rayon de courbure) contenue dans une base de données d'une voie ferroviaire ;
• la figure 4 est une vue schématique de l'enregistrement des données d'un capteur (ici, une accélération) lors d'un essai réalisé à une vitesse donnée sur la voie correspondante.
• la figure 5 montre en parallèle un enregistrement d'une grandeur, calculée à partir des valeurs de paramètres mesurés sur un véhicule en essai et l'évolution des imperfections en dressage de la voie, en fonction de l'odométrie de la voie.
• la figure 6 montre une analyse de l'amplitude des valeurs d'une grandeur représentative du comportement du véhicule en essai, en relation avec les imperfections en dressage de la voie sur laquelle celui-ci circule.

On these drawings:

• the figure 1 is a schematic representation of a vehicle under test on a railway,
• the figure 2 is a schematic representation, in plan view, of a railway track on which a qualification test is taking place.
• the figure 3 is a schematic representation, as a function of the odometric distance, of information (here, the radius of curvature) contained in a database of a railway track;
• the figure 4 is a schematic view of the recording of sensor data (here, acceleration) in a test run at a given speed on the corresponding channel.
• the figure 5 shows in parallel a record of a quantity, calculated from the values of parameters measured on a vehicle under test and the evolution of imperfections in dressage of the track, according to the odometry of the track.
• the figure 6 shows an analysis of the amplitude of the values of a magnitude representative of the behavior of the vehicle under test, in relation to the imperfections in dressage of the track on which it circulates.

Referring to the figure 1 a train is formed for evaluating the dynamic behavior of a railway vehicle 1. In this vehicle is installed a parameter recorder 2, connected to sensors placed in the vehicle, which measure values that may vary depending on the imperfections of the way. These recorded data are analyzed by an evaluation system 3 of the vehicle, which deduces from it quantities representative of the behavior of the vehicle and which compares these with permissible limit values.

The figure 2 represents a path consisting of a series of straight and curved sections, these two types of sections being connected by connection sections, not shown. The entry and exit coordinates of the curved sections are given by the kilometric distances Pki and Pki + 1 from an origin point O; the curves are described by their radius of curvature Ri.

Referring to the figure 3 we see the representation of one of the characteristics of a railway track (here, the radius of curvature) as a function of the odometric distance, as it comes from the database maintained by the track manager. To make this database, all channels are cut into elementary sections with homogeneous characteristics, which may be alignment sections, connection sections or circular curved sections. For each section the database provides information, among others, length of the section, through the odometric distances of the starting and finishing points of the section, and radius of curvature when it is curved. It also gives for the curved sections the value of the slope which is applied to the track to facilitate the curve keeping of the trains.

The figure 4 gives the results of recording a measurement parameter, when a test train is running on the corresponding channel. This parameter can be an acceleration experienced by a car or a bogie, a vibratory level, the force of support of a wheel on the rail or any other value dictated by the needs of the qualification sought. It is expressed as a function of the odometric distance Pk, placed on the abscissa and measured from the origin point O of the track. This representation is chosen to be homogeneous with the descriptive parameters of the track (radius, slope, ...) which are also expressed as a function of this odometric distance. The abscissa for each of the points of the curve is determined by transposing the recording, originally performed as a function of time, by an integration of the linear speed of the vehicle under test.

On the figure 5 two curves are represented, one giving, after the odometry has been readjusted, the evolution, along a particular section, of a quantity measured from a sensor (here the acceleration undergone) installed on the vehicle under test the other giving the evolution of the training of the track on this section, as it comes from the reference database. Dressing represents the difference, in the horizontal plane, of the axis of the track compared to the ideal layout. It constitutes an imperfection of the track, in the same way as defects of gauge of the rails or leveling of the way; all these imperfections are a priori listed in the reference database of the track.

The figure 6 is a representation of the magnitudes of these measured magnitudes, as a function of the amplitude of the erection imperfections noted on the track. It also indicates the permissible limit value for the quantity in question.

In order to be able to associate the measurement results with the track layout and to link the responses of the vehicle under test to the defects of the test vehicle, it is first of all important to readjust the arrival to a section with the reference distance of entry into the track. the section in the database. A measurement of distance calculated from a rev counter on an axle, associated with the measurement of the linear speed of the train, would not give a reliable result because of the phenomena of sliding of the wheel on the rail; a measurement of positioning by GPS would collide, in turn, tunnels and areas of darkness of this system.

To overcome these drawbacks it is proposed to add to the recordings made, an instantaneous measurement of the angular speed of yaw experienced by the car under test. This measurement can be carried out, for example, using a gyrometer. The measurement of the angular yaw rate is then related to the linear speed of the train to find the instantaneous value of the radius of curvature of the section on which the vehicle under test. Knowing the measured radius of curvature, it is possible to detect the moments when the radius of curvature changes or where it becomes constant, and thus to identify the odometric distances for which the vehicle has entered an aligned section or a curved section.

The comparison of the value of the radius of curvature with that theoretically provided by the database gives the difference d in odometric distance existing between the odometric values of the theoretical and actual entry (or exit) points of the various sections. This value d being, a priori, continuously known it is possible to perform a precise registration for each section and thus to analyze the records in relation to the database information by eliminating the errors due to the positioning of the mileage points. in the reference database. One can then see how the test vehicle reacts to the physical characteristics of the track and its local imperfections, as provided by the reference database.

It is thus possible to highlight, by means of the recording of a single angular velocity parameter, virtually all the errors and inaccuracies of the reference database and, once a registration has been carried out, to analyze in absolute terms the behavior of the vehicle, whatever the route assigned to the test and whatever the notice given for that test.

The invention relates to the analysis of the response in dynamic behavior of a railway vehicle to the various imperfections of the track. It is described here by taking the training of the track as an input parameter, but the process of qualifying a vehicle is concerned with all the types of defects listed in the reference databases of a track, and of which the examination is prescribed by the railway certification standards.

The first step is to redefine the division of the track into smaller sections than those in the reference database. These small sections, which are selected to conduct the evaluation, are defined by a standard for processing the stability of railway vehicles and will be called analysis sections thereafter. As an indication they are of a length of 70 or 100 m for the curved sections and 250 or 500 m for the sections in alignment, depending on the expected speed of movement of trains in commercial operation.

The next step is to classify the different analysis sections into statistical classes by grouping together, for example, all the curved sections that have radii of curvature between two specific values (such as a radius of curvature between 400 and 600 m). The stability analysis of the vehicle will then be carried out, statistical class by statistical class, and the vehicle will be declared good if it meets the requirements for each of the statistical classes.

For each analysis section and for each parameter of the dynamic behavior to be evaluated, a magnitude representative of the behavior of the vehicle is measured (an acceleration, in the case of FIG. figure 4 ), and we search on the analysis section considered if this value has reached a remarkable value in response to an imperfection of the way; as a remarkable value, it may be, for example, a maximum value or a peak on this value (as in the case of an acceleration shown in FIG. figure 4 ); it can also be the average value of the magnitude on the section, its standard deviation at a given percentile, etc. We also find the odometric distance for which this remarkable value is reached.

At the same time, the remarkable values of the imperfections of the track (the training in the case of figure 5 ) in the analysis section considered, and it is examined whether this remarkable value on the track can be associated with a remarkable value recorded on the measured quantity for the same section of analysis. For this purpose, these two values are only distinguished if the odometric distance separating them is less than a length agreed in advance; in practice, conventionally, this length is taken equal to half the length of the vehicle under test.

The process according to the invention ends with a graph of the type of figure 6 , which includes, on the abscissa, the remarkable values of the imperfections of the track for all the sections of the same statistical class and, on the ordinates, the remarkable values of the studied quantity which could be associated with these imperfections. These remarkable values are, finally, compared with a limit value defined by the calculation and represented on the figure 6 by a line of commonly encountered values, surrounded by a confidence interval, in which must be found the remarkable values of the parameter under consideration.

The qualification of the vehicle for operation at the speed at which the test was carried out and on a track with defects similar to those of the test track may be pronounced in respect of the parameter under consideration and the class section statistics considered, if no value of the dynamic characteristic examined exceeds a value prescribed by the standard. This examination can then be conducted for all statistical classes and for all the parameters imposed by the standard.

Although the invention has been described in connection with a particular embodiment, it is obvious that it includes all the technical equivalents of the means described and their combinations if they fall within the scope of the invention.

Claims (10)

Method for evaluating the conformity of the dynamic behavior of a railway vehicle (1) with respect to the stability rules of railway vehicles, by means of a test comprising a circulation of said vehicle on a railway track, said track being described in a reference database by indications on its geometry identified with reference to an odometric distance from an origin point, said database further comprising information on the imperfections of said pathway also identified by their odometric distance from the origin, characterized by the fact that it comprises the following steps: at. dividing said channel into analysis sections, b. classifying said analysis sections into statistical classes grouping the sections having similar geometrical characteristics, vs. recalibration by a known method of the odometric distances of the various analysis sections with the odometric distances of the sections contained in the database, d. measuring, when the vehicle is traveling on the track, at least one magnitude (γ) representative of the dynamic behavior of the vehicle, in response to the imperfections of the track, e. detachment of said quantity, in each of the analysis sections traversed by the vehicle, and extraction of the remarkable values of this quantity on each section, f. extraction, from the database, of the remarkable values of a characteristic (Dr) representative of the imperfections of the track, in each of the sections of analysis traversed by the vehicle, boy Wut. pairing, on each section, remarkable values of the quantity measured with those of the characteristic of the track, h. drawing a diagram giving the remarkable values of said quantity as a function of the remarkable values of the imperfection characteristic considered, for all the analysis sections belonging to the same statistical class, and i. comparison of the values obtained in this diagram with an admissible limit value. The method of claim 1 wherein the representative feature of the imperfections of the track is the training of said track. The method of claim 1 wherein the representative characteristic of the imperfections of the track is the spacing of the rails of said track. The method of claim 1 wherein the representative characteristic of the imperfections of the track is the leveling of said track. The method of claim 1 further comprising a step of repeating steps f to i for each characteristic representative of an imperfection of the path contained in the reference database. A method according to claim 5 further comprising performing steps a through j for a plurality of vehicle passage speeds (1) on the track and a step of determining the maximum speed allowed in commercial service on a track as a function of speed for the maximum value of the imperfections presented by said track corresponds to the attainment by a representative quantity of the behavior of the vehicle of its limit value. Method according to one of claims 1 to 6 wherein the division of the path into sections of analysis is performed as a function of the radius of curvature of said sections. The method of claim 7 wherein the statistical classes are defined at intervals of the value of the radius of curvature of said analysis sections. Method according to one of claims 1 to 6 wherein the division of the path into sections of analysis is performed as a function of the value of the cant of the track on said sections. The method of claim 9 wherein the statistical classes are defined at intervals of the value of the cant on said analysis sections.

Images